THEARUP JOURNAL

Vol. 8 No. 3 October 1973 Published by

Ova Arup Partnership

Sydney Opera House Special Issue

13 Fitzroy Street, London, W1 P 6BO Editor: Peter Haggett

Art Editor: Desmond Wyeth FSIA Editorial Assistant: David Brown

Contents Introduction, by J. Zunz Sydney Opera House, by Ova Arup and J. Zunz Design of the Concourse, by Ova Arup and R. Jenkins The glass walls, by D. Croft and J. Hooper Grouting prestressing ducts, by J. Nutt Adhesives for structural jointing, by T. O'Brien and J. Nutt Influence of corrosion on the design, by J. Nutt Sydney Opera House Awards, Credits Editor's note 2 4 22 30 42 48 52 54

Cover: East elevation of shell structure immense imaginative gifts. I n those early years he inspired all who came under his magic spell, and although there were great difficulties, they were gradually solved one by one and by 1963-64 the situation began to look quite hopeful. But then the going got rougher and Litton was pressed to produce drawings for the interiors. He didn't, couldn't, wouldn't, have it which way you will, and he resigned in 1966, leaving behind hard feelings, chaos, controversy, but above all a shattered dream. Whatever judgement posterity makes about Utzon's resignation and the subsequentfurore. no-one will deny his poetic, conceptual and visionary gifts and that his inability (for whatever reason) to complete the project is a tragedy. The truth is that he did walk out when information for the interiors and the glass walls was virtually non-existent. Hall, Todd and Littlemore were appointed by the New South Wales Government to the unenviable task of completing the job. They were faced with the now fixed parameters of the distinctive roof shape, with very definite accommodation requirements which could hardly be fitted in, and above all with a half-finished work of art - and Utzon's Opera House has an artistic quality with a capital A. Some of its critics have often said that there was too much art and too little commodity. Unfortunately, Utzon is not at the finishing post to prove whether they were right or wrong and half-finished works of art can never be wholly satisfactorily finished by others. Why did Utzon resign - did he jump or was he pushed? My guess is that he jumped. His behaviour, his letters, his interviews, all point to a path of self-destruction. He ditched his friends and collaborators for footling or no reasons at all and literally overnight left Australia never to return - at least not yet. Although Utzon's Opera House was the stuff that dreams are made of, although his use of shapes, materials, textures and colours was individual and introduced us to unique technical problems. I don't think that he ever really understood the complexity of the problems he was creating. Nor do I believe that he understood the problem-solving processes which ensued when new technology had to be developed or even when existing technology had to be adapted for new and untried forms. It is just possible that, in his seeming blindness to see that his collaboration with us was vital for the technical success of the scheme, lies another factor in his urge to leave the job. However, these are personal opinions. Despite the know-ails who have written and lectured on the subject, no-one will ever really find the truth. What is truth anyway? Whatever it is it will remain tucked away in men's minds. Post-Utzon, the affair became more orderly, though cost estimates still kept on rocketing, but control was a little tighter and problems became more easily soluble. However, the whole thing was none-the-less just a shade duller. What about Arups? What has the job done to us or for us, if anything ? Again, it is probably too early to see it in perspective, but there are some facts and some pointers. Firstly the facts -we stretched ourselves to the limits of our skills. In extending ourselves and making that extra effort we developed our know-howjust that little bit more. We use this knowledge in otherfields. When we have been extended as much as we have, it makes our ordinary jobs easier and we hope to do them better. We have had a good deal of publicity, some critical, but mostly complimentary: we have received the Queen's Award for Industry and, if travelling broadens the mind, many of us have had opportunities for mind -stretching. As for the more speculative consequences we were and still are in the middle of a great controversy. Our name is inextricably linked with the building. and while its success will be linked with Utzon and his successors, its failure will reflect on us. We became unwilling pawns in the controversy. On the one hand we wanted to help Utzon and do what was best for the job, on the other we wanted to act honourably towards a client who had treated us well and fairly. Whatever we did was bound not to please everybody. So we did all and sometimes more than was asked of us and what we thought was best for the job.

Introduction
Jack Zunz
For some years we have considered writing the Opera House story. We have never done anything about it-perhaps it was lack of time, will, motivation or even the doubt that the building would ever be completed. Now, nearly 15 years after construction commenced, the Queen will officially open it on 20 October. Instead of the book we didn'twrite we thought that the best thing to do would be to celebrate the end of the saga with a special issue of The Arup Journal. It contains some of the relatively few technical papers which have been written about the job as well as some selected photographs. It is difficult to believe that the festivities which will mark the opening ceremonies take place 16 years since we started work on the job. After the unending technical, human and political problems, after spending over 50m, we may well ask, was it all worthwhile? It is probably too early to say, but not too early to make some observations. Probably the most significant feature of the whole story is the astonishing reality that in a modern society, with all its checks and balances, its accountants and accountability, its budgets and budgetary controls, a folly on this scale could be contemplated. In other words, it is nothing short of miraculous that it happened at all. In concept it is not a building of this age. It has the romanticism of formereras when autocratic patronage made great follies possible. Yet, when Utzon's scheme was chosen from more than 200 competition entries, when the Premier of New South Wales was hell-bent on starting the job without drawings, and when all those associated with Utzon caught some of the euphoria of creating one of the great buildings of the age, it looked as though the improbable would come about after all. Much has been said and written about Utzon's 2 concept of the Opera House. He is a man of

The evolution and design of the Concourse at the Sydney Opera House
Ove Arup and Ronald Jenkins
This paper first appeared in the Proceedings of the Institution of Civil Engineers, April 7968. It is reproduced hereby kind permission of the Council of the Institution of Civil Engineers. The structure discussed in this paper generally known as 'the Concourse' or 'the folded slab forms a small part of the Sydney Opera House. The architect for the scheme was Jern Utzon, of Hellebaak, Denmark, and the structural engineers were Ova Arup & Partners, Consulting Engineers, London. The structure has a somewhat unusual shape, which was determined more by architectural than by structural considerations. As a rule authors of engineering papers only touch lightly on the development of a design and the aesthetic intention behind, it, confining themselves strictly to the structural, constructional and perhaps functional aspects, In the present case, however, the authors felt that this approach would be too narrow, because it would not explain why the structure was given this form. Certainly,functional and structural reasons alone would not have produced it, although they had a considerable influence on it. The first part of the paper will therefore try to explain how the design was produced by the joint effort of architect and engineers. Introduction Figs. 3 and 4 (p. 5) show two of the original competition drawings submitted by the architect, on which the location of the Concourse is indicated. In the following an attempt will be made to describe the progression of architectural and structural considerations put forward by the architect and the engineers which led to the chosen design. In Fig.4(p. 5) it will be seen that the architect had originally shown the Concourse supported on a number of columns at midspan. However, when this structure was first discussed between the architect and engineers, the architect asked whether it would not be possible to do without these columns. A typical question, which received the typical answer, that of course it was possible, but would cost a lot of money, and as the columns did not obstruct anything this expenditure might not be justified. The architect then explained that his concept demanded that the architecture should be expressed through the structure. in fact the structure in this case was the architecture; it should be bold, simple, on an impressive scale and of a form which combined sculptural quality with a clear expression of the forces acting on it. This achieved, finishes could be simple: the concrete itself would speak. The area covered by the Concourse was the place where people would arrive by car to the Opera House. and the impact of this vast unsupported roof would be spoilt by centre columns, even if they did not hinderthe traffic. Hefeltjustified in achieving the desired architectural effect by spending the money on a bolder structure rather than on expensive finishes. The solution proposed bythe engineers to meet these aspirations was based on the borehole data supplied by the client, according to which 22 firm sandstone would be found 3-4.6 in below

ground level, an assumption which much later was proved not to hold good for the crucial southern end of the site. It was also designed to solve the problem of draining this vast area (approximately 7000m2). The architect wanted the surface of the Concourse roof to be absolutely level, without the customary falls to drain off the water. Instead, the joints between the proposed 1.83x1.22m sandstone paving slabs would be left open to allow the water to seep through. This meant that it would not be necessary, and in fact not desirable, to provide a solid slab at the top; the supporting structure should be formed as,a series of channels leading the water towards the two ends of the Concourse but providing support for the sandstone paving slabs along line spaced 1.83m apart. Fig. 1 indicates the restrictions placed on the cross-section: (a) Support for paving slabs every 1.83m. (b) Channels in-.between (c) Total depth of structure should be uniform over the full length of the span, and this depth should be as'small as possible. Fig. 2 shows the longitudinal layout of one portion of the final structure.

There are two variants of this layout in other parts of the final structure, with different spans and different depth of structure, and before these dimensions could be fixed the design went through numerous variations, which however did not depart essentially from Fig. 2. It will be seen that if it is assumed that horizontal forces could be absorbed at A by the underlying sandstone, and at C by the substructure, which included a series of reinforced concrete boxes or longitudinal walls, then C-B and B-A could be strutted against each other, creating compression in both struts but reducing the moments. There would be maximum external negative moments at B and C and a maximum external positive moment somewhere in between. A glance at the shape of the 'arch' C-B-E shows that the angle CBA is critical; if it is too large the compression forces will be excessive, and the strains produced by these forces, by the stressing of the cables in these two members, and by creep and temperature stresses, could produce movements which might approach a critical stage. However, after the engineers' proposal had received the blessing of the architect, a preliminary investigation on the basis of the layout as it was then, proved that the proposal was structurally sound. Confining ourselves now to the main section C-B the task was to design a 'slab' or a series of beams, which would, as economically as possible, meetthe requirements in Fig. 1 which could take the negative moments at B and C and the positive at midspan, and which in a dramatic or sculptural way would reflect the variation in the external forces along the span and indicate how they were resisted at each point.

Support for paving slab every 6 ft

I
chann<t ro collet ninv,aco

is

is
f mlmum and uniform depth over full length of span

Fig.1 Basic requirements for cross-section

Max. positive moment E

Max. negative moment

Max. negative moment I e

Substructure

101 ft Fig. 2 Basic longitudinal layout

162 ft

Fig. 3 Cross-sections of various schemes (a)-(d)

This aspiration to have the structure 'truthfully displayed', to achieve 'structural honesty', is of course very familiar to students of architectural theory. It is a declared architectural ideal of long standing, and rightly so. But it must not be taken too literally. Geoffrey Scott showed 50 years agothatthis requirement was psychological rather than factual. It has nothing to do with choosing the most efficient structure. The spectator does not in fact understand the subtleties of a modern concrete structure, whose strength in any case may be hidden from the eye in the form of reinforcement or cables. It is not so much a question of how the structure really acts, but rather of how the spectator thinks it acts, or whether he can relate it to some simple structural facts which lie within his experience. Thus he may be able to appreciate the strength of an arch springing from solid abutments, a cantilever which is strong at its root, a simply-supported 'fishbelly' beam or a fixed beam with haunches producing an arching effect, and this may give him an impression of structural 'rightness'. More subtle effects would be lost on him; they would not form part of his architectural experience.
In this particular case the most economical answer would probably have been a series of box-sections or I-beams spaced 1.83m apart, uniform over the whole length, with prestressing cables catering for the variations in the moment. But this would obviously not have met the architect's request at all. It seemed natural to the engineers, therefore, to seek the solution by exploiting a typical and by now very familiar reinforced concrete form, the T-beam. This can be said to be the best shape to take positive moments in reinforced concrete. And the same shape, only upside-down, is the best shape for negative moments. In this way the desired expression of the variation in the external moments could be obtained by varying the shape from a series of inverted Tbeams at the supports to T-beams at midspan - or from section 1 to section 5 in Fig. 3. Such a solution would make structural sense in reinforced concrete if the formwork could be made reasonably simple, and full use made of the repetitive nature of the job. It would also be appropriate for prestressed concrete if the live load were small compared with the dead load, which was thought to be the case at the time. As it happened, due to the exigencies of the programme, it was not practical to place the paving slabs before prestressing, and these had therefore to be counted in with the live load, making the two about even. This considerably reduced the structural usefulness of the changing concrete sections, and made it impossible to justify the design on economic grounds. But it met the architectural requirements and there was no question of going back to a straightforward box-section.

d
(a)

(b)

(c)

(d)

t Fig. 4 Soffit of various schemes ( a)-(d)

The question then was howgradually to merge section 1 into section 5 (Fig, 3) in a manner which (a) Produced a sculpturally interesting soffit (b) Produced the lightest possible structure for the given depth, i.e. least redundant material (c) Was easy to construct. Fig. 3 shows four ways of doing this. In (a) the slab is simply raised through successive sectionsfrom 1 to 5. In (b) the walls are moved sidewaysfrom 1 through 2, 3,4and 5, gradually extending the top slab and contracting the bottom slab. In (c) the walls are gradually twisted, inclining more and more towards each other and reducing the area of the bottom slab until both slabs reach the same minimum, then twisted the other way, thereby increasing the area of the top slab until this covers the whole area, when section 5 is reached. In these three cases a further variable must be determined before the shape of the soffit is defined, namely the 'speed' of the change in

section along the axis of the beam. In order to make the shape as smooth and flowing as possible the engineers decided to make the change in section follow a sinusoidal variation. In scheme (a) this form of variation would produce a series of beams 1.83m apart connected by a wavy slab; in (b) there would be 'wavy' beams connected in alternate bays by flat top and bottom slabs, and in (c) the wavy beams would be twisted at the same time. In Fig.4 (a), (b), (c) and (d) are attemptstoshow how the soffits would appear. In the engineers' opinion there was a progression from (a) to (c) in aesthestic interest and also in some ways in structural suitability, but unfortunately the complexity of formwork was also increasing. However, as the twisted surfaces of (c) contained straight lines, these could be produced easily enough from straight boards or twisted plywood. This was therefore the solution put forward to the architect, but in a slightly modified form, as indicated in Fig. 3 (d). It seemed to the engineers that scheme (c), seen from below, looked too much like a flat soffit with certain regular hollows scooped out of it. By connecting the hollows together, i.e. by introducing a piece of top slab between the 'beams' of the same width as the beam at midspan, the appearance was more of a series of swelling and undulating beams, and the shape of the beam soffits was repeated in the under-

side of the top slab. Another minor modification was that the beam sides, instead of being vertical in sections 1 and 5, were slightly slanted to facilitate withdrawal of the formwork. Scheme (d) was at once approved by the architect, and was the one incorporated in the preliminary design submitted to the client by the architect and engineers in April 1958, and approved. The architect had however introduced a further modification in the design, which the engineers were not too happyabout. He insisted that the visible corners between the modulating walls or beams and the soffits of top slab and beam should be rounded off, as shown in Fig. 5, and explained that this was very necessary in order to bring out the sculptural quality of the design. The engineers did not dispute this, but were worried about how to producethese rounded corners, and thought it would be very difficult and expensive. It had been their idea that the forms should be made of straight narrow boards forming the twisted surfaces, which would therefore show the familiar boardmarkings characteristic of structural concrete. However, the architect demonstrated on a small model that these board marks and the sharp corners would be out of scale, and that the desired effect could only be achieved by smooth rounded surfaces. Ajointvisit paid bythe architect and the engineers to the Sydney plywood factory of Messrs. Symonds, who were masters in the manipulation of plywood, confirmed that the architect's ideas would be difficult to realize, and on the return journ ey from Sydney the designer therefore considered other and more practical ways of effecting the transition from section 1 to section 5 in Fig. 3 (d). It appeared that there were not so many simple ways of effecting this transition, if one observed the rule that the cross-sections should always be made up of straight lines, which would then produce twisted surfaces which could be made up of plywood. The method proposed in Fig. 3 (c) and (d) seemed to bethe simplest possible, i.e. tilting the side B-C (Fig. 12, p.7), rotating it round point B until point C coincided with D, 23

Fig.5 Proposed round-edged sections

then twisting the side back in the other direction, rotating about D until point B coincided with F. The next simple method (Fig. 6) seemed to be to rotate the side B-C round point B as before, and simultaneously to rotate part of the beam soffit D-C round D, in such a way that the point of intersection C between the two lines moved on a straight line from C to its ultimate destination, point F. The resulting shape of the beam, assuming that the sinusoidal variation of cross-sections was maintained, proved to be very interesting and to possess that roundness or voluptuousness which the architect was looking for, in spite of the fact that there were no rounded corners. Fig. 13 (p.8) shows some typical crosssections and Fig. 14 (p.8) a dimensioned section of the executed scheme. After considering this new proposal and making models to judge its effect, the architect wholeheartedly approved of it, adopted it, and had it passed by the Technical Panel. The engineers, having concentrated their attention on obtaining an architecturallyinteresting solution which could be produced with fairly simple formwork, had at that stage possibly given too little weight to one possible disadvantage of the last scheme compared with that in Fig. 3, namely that the 'kink' in the side walls might increase the internal bursting stresses produced by the bending stresses resulting in an increase of ordinary reinforcement, thereby adding to the difficulties of compacting the concrete. Butthiswas only a minor snag compared with many others which emerged during the detailed design and the execution of this work. For one thing, the assumptions on which the design was based underwent various changes, all for the worse. It was found, for instance, that the underlying sandstone dipped down at the southern end of the site, making it doubtful whether the safe bearing which the design called for could be provided at this end. Then the architect changed, at the Technical Panel's request, the ratio between treads and risers of the steps, flattening the slope of A-B (Fig. 2). As pointed out earlier, the angle C-B-A had a vital influence on the horizontal forces which had to be absorbed. For these reasons it was found desirable to introduce tie-beams between the foundations atA and C. This added to the cost, but put the design on a much sounder basis, and part of the cost was offset by the fact that the prestress produced by the cables in the tiebeams made it possible to reduce the number of cables in the folded slab itself. All these changes naturally delayed the completion of the detailed drawings which were urgently needed on site, and further aggravated the almost impossible situation which was created by the client's insistence that work should begin on site early in 1959, long before the brief - let alone any finished and dimensioned drawings - had been completed. The situation was not improved by the contractor's insistence that his programme demanded an early start on exactly this particular part of the job. Add to this the difficult nature of the job, complicated or unusual formwork, narrow sections packed with steel, etc., and the contractor's unfamiliarity with prestressed concrete. and it is no wonderthat the atmosphere on the job deteriorated and the workmanship suffered. A description of the snags which developed and of how they were overcome would perhaps be useful but falls outside the scope of this paper. But it may be of interest to mention another complication which was happily avoided, because it concerns the design, and it throws some light on the somewhat different points of view of architect and engineers. It arose from the fact that a part of the Concourse slab (the part and er the restaurant) was raised a few steps over the rest. It was part of the architect's philosophy - to use a now 24 popular phrase - that the structure, i.e. the

Fig. 6 Isometric view of executed scheme

Fig. 7

Cross-section through proposed slab

Fig. 8 Cross-section through executed slab

shape of the slab as seen from below, should register this fact: one should be aware of what happened above, just as one should be aware of the forces acting on the slab. The crosssection in Fig. 7 shows what the architect wanted to do: the beams under the higher portion are lifted up and there is a gradual transition to the normal level, more or less following the steps above. It was difficult to argue that this could not be done, although it posed tremendous problems, because the five special beams were unsymmetrical in crosssection and the prestressing would create torsional movements which would have to be absorbed by the adjoining, already fullystressed, beams. This would require structural additions and might even prove to be almost impossible - apart from the fact that it would upset the whole arrangement of stressing two adjoining beams at a time, and would require five sets of special and more complicated forms, thereby invalidating the excuse of repetitive formwork. The engineers' view was that even if the architect was correct in preferring his solution from an aesthetic point of view-which they did not dispute - the very considerable cost, and the disturbance itwould cause in an alreadycritical situation, would be too high a price to pay for something which after all would not be missed

by anybody. However, the architect was insistentand the engineers were bracing themselves to attempt a solution to the problem when the heating engineers intervenedwith ademandfor space overtheslab in whichthey could accommodate their pipes and other services. This clinched the matter: by keeping the beams at the same level as in Fig. 8, the desired space would automatically be created and everybody was satisfied! Figs. 9 and 10 show the appearance of the Concourse slab from underneath and from the side, and Fig. 11 is an aerial view of thefinish ed slab without the pavement. This account of the development of the design is necessarily brief and deals only with the typical case. Amongst otherthings it leaves out the considerable difficulties in creating the correct boundary conditions, especially for the cross-beams spanning the openings in the supporting wall at the north end, but these are dealt with in the next part of the paper. Structural design The Concourse is a folded-plate structure in prestressed concrete and the analysis was quite conventional. Some unusual features were present, however, because it was not simply a bridge but had also to have the architectural attributes already described.

The depth/span ratios were low and the shape was not ideally suited to prestressing. The structural design was. thus, mainly concerned with keeping withintheworking stresses under all conditions. The cross-sections continuously varied according to geometrical rules. For this main reason most of the numerical work was programmed for the electronic digital computer. The analysis was done and the working drawings prepared in 1959. The remarkable feature was the extremely flat angle of 191 of the leg from A to B (Fig. 11, p.7 ). This madethe portal part of the structure (A, B, C), very sensitive to the phenomena of concrete movements. Time, temperature and load gave many combinations to investigate and reinforced the case for computer working. A general description of the whole structure will now be given. The part denoted by RC substructure in Figs. 10 and 11 (p.7) consisted of two-storey boxlike structures which received the prestressed tie-beams in a floor at that level. The compression in the tied portal due to the thrust at the foot, A, was taken up at the other end by shear walls in the substructure. The couple thus imposed on the substructure required special measures forstability, but was a greatly reduced problem compared with the original idea of thrusting against the rock, because the indicated rock level was well below the tie level. The substructure was not there just for the structural purpose described. The shear-wall spacings were determined by the various uses of the rooms. A certain amount of structural irregularity was a small thing compared with the large spaces required for main stairway entrances and especially for the service road, 1 2.8 m wide, which entered the Opera House attie-beam level atthecentreof the Concourse. These boundary conditions at the north side of the portal are shown diagrammatically in Fig. 10 (p.7). The upper reaches of the Concourse shown typically from C, D to E in Fig. 11 (p.7), were subject to several varieties of spans, slopes and flats. The figure indicates that where visible to the public, the architect required the 1.83m wide, varying section, folded slabs to flow continuously into the upper reaches. Because of the reduced depth the most critical point was found in the upper sloping part. For an ordinary continuous beam one would have introduced something like a concrete hinge overthe intermediate support at C. However, the vertical reaction at C was combined with a horizontal reaction of the order of 203 tonnes/1.83m wide Concourse beam. A hinge under these conditions would have required costly mechanical devices. It was decided to make the beams monolithic with the shear walls, which in the region of C had thus to receive the vertical reaction, the portal compression and a couple equal to the difference between the end moment of the portal part and that of the upper part of the Concourse. Transom beams, in one case of considerable size.were introduced acrossthewide openings mentioned above. The transoms were in a state of vertical and horizontal bending and torsion. The torsion arose from the beam end moment differences which depended on the combinations of temperature and loading. The minimum would be obtained if maximum clockwise and anticlockwise moment differences were numerically equal. By preloading and other devices the upper spans were made as far as possible to bring about this optimum condition. The stresses in the transoms resulted in deflections and rotations, but these were of a small enough order not to influence the assumed fixed end condition of the Concourse. One meets a parallel case in the edge beam of a cylindrical shell. When the beam is of normal size it makes no material difference if one takes

into account its torsional rigidity or assumes it does not twist. The portal part of the Concourse contained 47 folded slab units, 1.83 an wide. The slabs were 178mm thick: therewere 21 units 50m in span and 1.37m deep; the remainder were 41.5m in span and 1.14m deep. Both types were designed in a similar way. The longer span has been selected for description in the paper. With the sloping leg at such a flat angle, a structure with unusual sensitivity to concrete movements seemed to provide a good opportunity for correlating calculated and measured strains and deflections. The distance from London, where the design was done, made the site measurements less extensive and accurate thanwas desirable forthe exercise. The authors do not believe that the results add anything to present knowledge (a common finding unless the instrumentation is good) and the correlation will not be given in this paper. Long-term movements were important. The construction of the portals was begun in November 1960, and finished in January 1963. The laying of paving slabs, which are a permanent superimposed load, may not commence until 1968/9. The application of full live load will be a rare and short-term event. Ultimately the concrete of the portals will be shielded from direct sunlight by the paving. The average temperature condition will be when the temperature of the portal concrete is the same as that of the tie concrete. Long-term factors are that the paving should be quite flat at average temperature when there are few people on it and the possibility of further creep dueto changes of stress from the weight of the paving.

The control devices used to make these structures largely independent of concrete movements not accurately known will be described. However, for the sake of the correlation the designers made some research into published experimental information and made use of what they thought was the latest at the time, not very different from that given in the British Standard Code of Practice (CP115:1959). A pair of connected folded slab units 3.66m wide, were cast at a time. A gap, to be filled in later, was left between one pair and the next pair. In this way stressing operations could be carried out on a pair of connected units without disturbing the adjacent units. The pairof ties corresponding to a pairof folded slab beams passed on opposite sides of the T-sectionsatAand were connected bya crosshead. Hydraulic ship jacks of 203 tonnes pressure were introduced into the gap between the cross-head and the foot of the portal. Two ship jacks were used and oneotherwas keptas a standby in case of breakdown. Also in the gap was a pair of steel wedge assemblies. The details of the jack and wedge apparatus are shown in Figs. 11 (p.7) and Fig. 15 (p.8). The former also shows the profiles of the 28.6 mm diameter strand post-tensioning cables. Load cells or dynamometers were used with the hydraulic jacks and the prestressing jacks so as to obtain more accurate force measurements than could be relied upon from pressure gauges. The order of prestressing and jacking was carefully worked out so that the structure could be converted from an unstressed, inert state supported by soffit props to a fullystressed free-standing condition without at any stage exceeding the permissible design stresses attransfer. Creep and shrinkage losses in the portals were made good by periodic

Fig. 9 Completed structure 1

Fig.10 Completed structure 2

F1g.11 Aerial photograph of site showing Concourse area

25

rejacking, which meant moving the ship jacks around quite a lot. The folded slabs were cast to true final shape. The knee at B could be maintained to correct height by the ship jacking and geometrical non-linearity could thus be avoided. The jacking was not simplyto allow for concrete movements. The calculated force obtained optimum stress conditions in the portals. The jacks were, in fact, a second method of prestressing. The tie-beam had been previously prestressed to a concrete compression of about 9 N/mm2. Thus the operation of the jacks decompressed the concrete. Data of creep recovery for the intended correlation were found to be too scanty for consideration. A case could be made for the use of Freyssinet flat jacks throughout in place of the ship jacks and wedges. Th e ship jacks were ordered when it was thought the jacking might be against rock instead of ties. The jack travel might then have been quite large to an extent that could not be predetermined. To deal with the possibility of extra creep due to the paving load the intention is to stack the paving slabs in transverse lines across the Concourse. The jacks will then be put into position again and used to compensatefor any loss of thrust and maintain the correct level at B. The jacking gap will be concreted in when everything has settled down and the gap left between every pair of units will be made good. The paving will then be laid true to level at a period of average temperature. It should be mentioned that since there was an odd number of folded slab units of the long span type, one unit had to be made by itself with a gap on each side. Materials Concrete The fine aggregate was a uniformly-graded Cronulla sand with approximately 4 per cent moisture content. The coarse aggregate was Prospect Blue Metal, a crusher run aggregate, uniformly graded with a maximum size of 22 mm. A concrete mix was designed whose proportion by weight was 1 :1.19:2.43. The water/cement ratio was 0.39. An additive, Darex WRDA, was included at the rate of 2.2 kg/ma of concrete mixed. Theauthors' specification called fora minimum cube strength of 48.3N/mm2 at 28 days, and prestressing was allowed to begin when field cubes reached an average strength of 41.4 N/mm2. The permissible working concrete stresses adopted in design, were: N/mm2 1 5.5 Bending compression Bending tension Shear stress Local bond stress Average bond stress Principal tensile stress 1.5 1.2 1.2 0.8 1.1
V, UNKNOWN T

Fig. 12 Construction sequence

housed in Kopex ducting with an inside diameter of 38 mm. The mechanical properties of the strand, assumed in design were: Young's modulus, tangent 1 65 kN/mm2 Secant modulus at 63 tonnes Friction constant between cable and ducting 131 kN/mm2 p=0.30

Wobble constant K=0.0010 The use of the calibrated dynamometers at the live end and dead end of several cablesshowed K=0.0013 and p=0.25,. It was found that the change of these constants from those assumed did not materially affect the analysis. Loss of prestress The relaxation of the strand cables was determined by experiments carried out at the University of Sydney. Loss of prestress partly depended on the distance from anchorages, the angle turned through and the concrete stress at the centroid of the cables, under dead load conditionswith-

out paving. The loss at transfer was distinguished from the ultimate loss after several years under dead load. The ultimate loss averaged about 33 per cent. The approximate average contributions were 19 per cent due to friction, 5 per cent due to creep and shrinkage, 5 per cent due to anchorage slip and elastic shortening and 4 per cent due to relaxation of the strand. The latter was reduced from 6 per cent by holding the maximum force of 63 tonnes for five minutes before wedging-off. The high friction loss was partly due to the large proportion of cables stressed from one end only through restrictions in fitting stressing jacks into position at both ends. The restrictions arose from having all theanchorage in the upper hollowed-out part of the beams, for the visible undersides could not be marred by the making good of anchorage pockets. The elastic shortening mainly arose from the thrust of the ship jacks. The prestressing cables were grouted before the ship jacks were brought into full operation. Thermometers Thetemperature difference between the folded

c-FOLDED SLAB CENTROID

0
I H . K NOWN

The prestressing strand was augmented with mild steel reinforcement where tensile stresses were found. The coefficients assumed for calculating deflections were: Young's modulus at transfer 34.5 kN/mmz Long-term Young's modulus 17.2 kN/mm2 Young's modulus in tie 34.5 kN/mm2 Shrinkage strain 0.03% Prestressing The prestressing equipment for 28.6 mm diameterstrandwas by Gifford-Udal and a dynamometer of Swedish construction was added for accurate measurement of stress. The 28.6 mm diameter prestressing strand, which had then been only recently marketed, had a guaranteed 26 minimum tensile strength of 82 tonnes. It was

Fig. 13 Diagram for analysis with one unknown

FOLDED SLAB CENTROID

^71E

H
UNKNOWN

o
V,UNKNOWN t i

Fig. 14 Diagram for analysis with two unknowns

tR 0

30

sso 10 Fig. 15 Relaxation/time curves s o

10
NUMBER OF DAYS AFTER PRESTRESSING C LOS SCALE I

100

Fig. 16 Diagram for structural analyses

400

N a_

FOLDED SLAB COLDER THAN TIE BEAM

- 350

E is constant throughout, which would be the case if creep is proportional to stress, it does PArR not enter into load analysis because its value sue WL, S T'. comes in the denominator on both sides of the equation of relative movements due to loads in v raceo and those due to the unknowns (redundants). In fact, influence coefficient methods amount to finding the redundants by this equation. If there is more than one redundant, the equations are simultaneous, as in the second case. When there was a relative temperature change while the horizontal thrust was held on the wedges, the second case of two unknowns lac , arose. Further, in this case there were relative movements in the released system which comprised the change of vertical height of the sloping leg and the relative changes of horizontal length of folded slab to the tie-beam. which had to be eliminated to restore continuity by redundant actions, It is therefore evident that the value of Ecomes into only one side of this equation and can not be cancelled. The folded slabs were exposed to the air and the tie beams were not. The daily alterations of temperature were small, but there were sometimes larger of the same sign over pee ods of severe le days to a week or more. The temperature differentials were regarded as short-term events, and the E value used was 34.5 kN/mm2. Thus, in the judgement of the designers, creep due to temperature stresses was ignored. As regards the application of the permanent paving load, the structure will still be under the wedge and jack apparatus, and any further creep will still betaken up. If it is presumed that creep will take place where the compressive stress is increased but notwhere it is decreased, the creep will be resisted by peculiar stress distributions overthe cross-sections, unless the creep recovery factor is equal to that of creep strain which seems unlikely. These are stress history matters on which published research was entirely lacking. The loadings taken for design were: kg/m per unit FOLDED SLAB Dead load or self WARMER THAN weight 2315 x area of section 27
TIE BEAM

x r

Paving Live In plan these are: Dead load Paving Live

300

= 171 kg = 272 kg
kg/m2 = 1025 average = 308

= 488

Fig.17 Gain or loss of thrust due to temperature

-25

-10 +10 t0 TEMPERATURE DIFFERENTIAL IN DEGREE FAHRENHEIT

.25

slab and the tie had an influence, as will be explained, on the forces applied by the ship jacks. Eight thermometers were located in each pair of tie beams. The thermometers were immersed in water in specially constructed temperature pockets and read just before the jacking operation. The temperature range used in the structural design was 14C. Portal design The portal foot at A rested on a foundation through lubricated sliding plates. Two cases of statically indeterminate structures were analysed by the influence coefficient method. The analyses were programmed forthe electronic digital computer to eight significant figures. In view of this the simplest released system was adopted, that of cantilevering the whole folded slab structure from C. Bending and direct strain were taken into account, but not shear strain. However, the program was arranged to give shear forces. Part of the data was the prestressing tensions and their slope, which were taken into account for resultant shear on the concrete.

Thefirst case is shown diagrammatically in Fig. 13. At A, the horizontal thrust is known and the vertical reaction is the unknown. The second case is shown in Fig. 14.The unknowns arethe vertical and horizontal reactions at A. The designers had to consider the effects of concrete movements on these indeterminate structures. In the first case, the horizontal thrustwas to be brought up from time to time to a known force over a period of several years. There was no doubt that shrinkage, creep, elastic tie extension, and the loss of the horizontal component of folded slab compression would be taken up by jack travel, The vertical component of the compression in the sloping leg, naturally, came into the analysis. When it came to rejacking, it was found, as expected, that when the thrust was just taken up there had been a loss. Fig. 15 shows two plots for loss of initial thrust against days on a logarithmic scale. The differences in the curves were due to differences in time from casting the concrete to stressing operations. Itis well known that if the modulus of elasticity

Integrations for the analysis were performed numerically in the computer by summing according to repeated Simpson's rule, These were both total integrals over the whole structure and integrals from A to the interval points. Cross-sections were drawn to scale at each interval (numbered 0-27 on Fig. 16), to provide dimensions for the two purposes of making the formwork and a computer run to give the area A, the moment of inertia / and the distance of the centroid. { in the figure, from the straight reference lines, AB, BC, at each cross-section. Due to certain changes afterthe cross-sections had been drawn, some unequal intervals appeared, which required modificationtotheSimpson coefficients in their regions. From this preliminary program, the co-ordinates 'of the true axial line (centroid) at each interval were given, denoted by y, z in Fig. 16. The graph used on the site to determine the appropriate jacking force at various temperature differentials is shown in Fig. 17. Computer runs, for the two spans, were made for 19 conditions. These comprised transfer, self-weight and paving, and live load; maximum and minimum temperature differences; and jack thrusts ranging from 203 to 376 tonnes. The latter were for determining how much loss of thrust could be tolerated and to what extent over-jacking could be employed to compensate for future losses. 27

SS pp su,ydwoy g uosdwogl Aq puel6u3 ul pO UIld

(uopmo$ AmH ologd)

ZA/

!' ^
`lnu""4150^ kM ^

P e^aARO

- ^.
^^^^-1111au